Repository navigation
Explain how agentic AI fits into the Software Development Model - #35
Merged
Merged
Conversation
Agents are strong at turning an agreed decision into an artefact and weak at reaching the decision. In the model, that is the split between component-to-component relationships and the relationships from the Domain Expert and Software Engineer into components. Their collaboration stays; the central tension is speed against shared understanding. The explanation walks the five focus areas in flow order, then the tensions across the model and the limits of the evidence. It rests on new research for Align and Understand, BDD, DDD and Lean that reuses the XP research: 60 sources, average reputation 0.89. Both passed peer review. Readability measures about 55 against a 70-80 target; accepted, as the gap comes from glossary terms that must be used exactly. The mapping is interpretation only; the model is unchanged. Refs #34 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
ibanFR
changed the base branch from
31-research-agentic-development-in-xp
to
main
September 30, 2026 18:44
There was a problem hiding this comment.
Copilot review overview
🟡 Changes recommended
Several claims lack direct citations, and the source-validation record contains unresolved inaccuracies.
Review effort: Balanced
Findings: 2
Open (10)
What changed in this PR
Explains how agentic AI supports—not replaces—the Software Development Model’s roles and focus areas.
Changes:
- Adds the evidence-backed explanation.
- Adds supporting research and citations.
- Records documentation validation and peer-review results.
| File | Description |
|---|---|
docs/explanation/agentic-ai-in-the-software-development-model.md |
Explains agentic AI’s role across the model. |
docs/explanation/agentic-ai-in-the-software-development-model.md.validation.yaml |
Records validation and review outcomes. |
docs/research/agentic-ai-in-the-software-development-model-for-explanation-doc.md |
Provides supporting research and source analysis. |
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
Link the QCon report, the Siddeeq Gherkin study, the domain-modelling benchmark, the 2x-mandate field study, the feedback-classification study, the skill-formation studies and Farley's podcast, and separate what each user-story study found instead of attributing all three weaknesses to both. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Copilot review overview
🟡 Changes recommended
Several evidence attributions, model descriptions, citations, and validation entries remain inaccurate or inconsistent.
Review effort: Balanced
Findings: 4
Open (8)
Resolved since last review (10)
- Describe BDD and DDD as separate paths from Align and Understand that each feed XP, as workspace.dsl shows. - Attribute each Gherkin finding to its own study instead of all three. - Cite the 2025 DORA report and the AI Capabilities Model. - List Strategic Architecture, not the collaborative modelling relationship, among the collaborative components. - Relabel the research review entry: N2 and N3 are BDD studies. - Fix Plöd's name and bring the validation record in line with the measured readability score. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes #34
Stacked on #32: the explanation links to the XP research added there. Merge #32 first; this PR then retargets to
main.Adds
docs/explanation/agentic-ai-in-the-software-development-model.md, its validation file, and the research behind it,docs/research/agentic-ai-in-the-software-development-model-for-explanation-doc.md.Answer in short: agents are good at turning an agreed decision into an artefact and weak at reaching the decision.
The doc walks the five focus areas in flow order, then the tensions across the model and the limits of the evidence. The mapping is interpretation only:
workspace.dsl,CONTEXT.mdand the ADRs are unchanged.Sources: 60, 26 new and 34 reused from the XP research, average reputation 0.89. Research review and doc review both approved on the first pass; their results are recorded in the research document and the validation file.
Weak spots:
🤖 Generated with Claude Code